Benchmarking

SNF data description

• The SNF datasets were part of the datasets used in the Nature Methods paper on Similarity Network Fusion (SNF); https://www.nature.com/articles/nmeth.2810
• The cancer datasets include GBM (Brain), Colon, Kidney, Lung and Breast (the Breast cancer dataset was excluded in order to avoid confusion with the case study on Breast Cancer)
• The datasets were obtained from: http://compbio.cs.toronto.edu/SNF/SNF/Software.html
• Survival times were provided for each disease cohort. The median survival time was used to dictomize each response variables into low and high survival times.

number of samples in each group

##      colon kidney gbm lung Sum
## high    33     61 105   53 252
## low     59     61 108   53 281
## Sum     92    122 213  106 533

number of variables in each dataset

• mRNA transcripts or cpg probes that mapped to the same gene were averaged

##       colon kidney   gbm  lung
## mrna  17814  17665 12042 12042
## mirna   312    329   534   352
## cpg   23088  24960  1305 23074

Multi-omic biomarker panels

Unsupervised

JIVE

MOFA

sGCCA

Supervised

Concatenation_sPLSDA

Ensemble_spslda

DIABLO_null

DIABLO_full

Number of features per panel

Component plots

Overlap in panels

Colon

Intersection plot

Venn diagram

Kidney

Intersection plot

Venn diagram

GBM

Intersection plot

Venn diagram

Lung

Intersection plot

Venn diagram

Gene set enrichment analysis

• We wished to assess the enrichment of the selected features across a variety of annotated gene sets in the MSigDB collection (http://software.broadinstitute.org/gsea/msigdb), in particular:

1. C1 - positional gene sets for each human chromosome and cytogenetic band.
2. C2 – curated gene sets (Pathway Interaction DB [PID], Biocarta [BIOCARTA], Kyoto Encyclopedia of Genes and Genomes [KEGG], Reactome [REACTOME], and others)
3. C3 - motif gene sets based on conserved cis-regulatory motifs from a comparative analysis of the human, mouse, rat, and dog genomes.
4. C4 – computational gene sets (from the Cancer Gene Neighbourhoods [CGN] and Cancer Modules [CM] – citation available via: http://www.broadinstitute.org/gsea/msigdb/collections.jsp)
5. C5 - GO gene sets consist of genes annotated by the same GO terms.
6. C6 – ontologic gene sets (Gene sets represent signatures of cellular pathways which are often dis-regulated in cancer).
7. C7 - immunologic gene sets defined directly from microarray gene expression data from immunologic studies.
8. H - hallmark gene sets are coherently expressed signatures derived by aggregating many MSigDB gene sets to represent well-defined biological states or processes. & A. BTM - Blood Transcriptional Modules (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2727981/) B. TISSUES - cell-specific expression from Benita et al. Blood 2008 http://www.bloodjournal.org/content/115/26/5376

• Significance of enrichment was determined using a hypergeometric test of the overlap between the selected features (mapped to official HUGO gene symbols or official miRNA symbols) and the various gene sets contained in the collections. Resulting p-values were corrected for multiple hypothesis using the Benjamini-Hochberg procedure applied across ALL genesets (10k+ tests – as pessimistic as possible). Adjusted p-values are reported in the fdr column.

mRNA and CpGs

## # A tibble: 28 x 13
##    disease        method         type   BTM    C1    C2    C3    C4    C5
##  *   <chr>         <chr>        <chr> <int> <int> <int> <int> <int> <int>
##  1   colon Concatenation   supervised     0     0    12    11     1     7
##  2   colon   DIABLO_full   supervised    23     0   113     0    46   216
##  3   colon   DIABLO_null   supervised     0     0    21     6     1     0
##  4   colon      Ensemble   supervised     0     0     3     2     2     0
##  5   colon          JIVE unsupervised     0     0    15     8     0    19
##  6   colon          MOFA unsupervised     4     0    14     5     1    36
##  7   colon         sGCCA unsupervised     0     0     5    14     0   147
##  8     gbm Concatenation   supervised    10     0   258    14    47   526
##  9     gbm   DIABLO_full   supervised    30     0   426    34   125   693
## 10     gbm   DIABLO_null   supervised    10     0   312    15    62   776
## # ... with 18 more rows, and 4 more variables: C6 <int>, C7 <int>,
## #   H <int>, TISSUES <int>

which method is leads to the greatest number of signficant pathways?

Connectivity

datasets of multi-omic panels

## [1] "JIVE"          "MOFA"          "sGCCA"         "Concatenation"
## [5] "Ensemble"      "DIABLO_null"   "DIABLO_full"

adjacency matrices

Number of connections

Network attributes

all network plots

network and component plot of the multi-omic panels derived using the colon cancer dataset

networks

colon component plot